Re: envsys/gttwsi deadlock

[Taylor R Campbell <campbell+netbsd-tech-kern%mumble.net@localhost>]

   Date: Wed, 14 Oct 2015 12:35:41 +0200
   From: Manuel Bouyer <bouyer%antioche.eu.org@localhost>

   A way to fix this would be to use mutex_tryenter() in sme_events_check(),
   and just reschedule the callout if it can't get the lock.

   Does anyone see a problem with this approach ?

That may be a temporary stop-gap, but it is wrong to sleep in
gttwsi_wait while holding sme->sme_mtx, or any other mutex (e.g.,
sme_global_mtx).

This is not gttwsi_wait's fault -- this is envsys's fault.  The
correct solution should be to change the API to avoid holding a mutex
across calls to a device sensor refresh routine.  Something like
kern/kern_rndsink.c, perhaps.